|Publication number||US7269648 B1|
|Application number||US 09/964,977|
|Publication date||Sep 11, 2007|
|Filing date||Sep 27, 2001|
|Priority date||Sep 27, 2001|
|Also published as||US7730180|
|Publication number||09964977, 964977, US 7269648 B1, US 7269648B1, US-B1-7269648, US7269648 B1, US7269648B1|
|Inventors||Sriram Krishnan, Gregory W. Lazar|
|Original Assignee||Emc Corporation|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (52), Referenced by (51), Classifications (6), Legal Events (5)|
|External Links: USPTO, USPTO Assignment, Espacenet|
This patent application relates to another patent application filed on even date herewith having co-inventors Sriram Krishnan, Andreas L. Bauer, and Gregory W. Lazar, assignee in common with this patent application, the title of “Managing a Distributed Directory Database”, and which is incorporated by reference herein in its entirety.
1. Field of the Invention
The present invention relates to apparatus, methodology, systems, and/or computer program product for resolving multiple master conflict in a directory database (DDB) and, more particularly, relates to resolving multiple master conflict in a globally-distributed DDB within a computer network environment such as a client server network.
2. Description of Prior Art
As may be observed by even the most casual computer system user, computer networks appear to be continuously evolving into ever-more sophisticated communication systems. Accordingly, computer networks are having an ever-increasing impact on modern-day lifestyle for many people, as the Internet is demonstrating. Computer networks include local area networks (LANs) such as might be encountered within a singular and localized corporate business organization where corporate functions (engineering, marketing, sales, advertising, human resources, etc.) are linked together in a localized computer network. Computer networks also include wide area networks (WANs) such as might be encountered within a corporate business organization that is nationwide or even worldwide in scope. The Internet is the premier example of a WAN, but is one that is obviously not constrained within any particular business organization and allows access to virtually any organization or individual. It is widely understood that these networks not only utilize hard-wire bus communication paths to accomplish their objectives, but can also utilize wireless telecommunication links via satellites and the like as well.
Computer network configurations usable within LANs and WANs include client-server network configurations. A client or workstation (typically having a user interface) can be networked to multiple servers which serve the client in a variety of ways. Such a client-server computer network configuration can be particularly useful in certain applications, such as in managing a computer data storage system. A client-server network can be used with the Internet. Thus, quality of operation of a storage system within a client-server network operatively coupled over the Internet not only can impact quality of performance of that client-server computer network, but can also impact quality of service received by multiple human users connected through the Internet to that client-server network. A failed or degraded storage system thus can escalate into a failed or degraded service for multiple human users.
A client-server network today may typically be based on an object oriented computer system which means that such system employs one or more object-oriented computer languages such as C++, XML (eXtensible Markup Language), JAVA, and/or others. Briefly, an object, in computer software terms, also referred to as a “node”, is a dedicated area of memory which can be thought of as an impervious container holding both data and instructions within itself, both defining itself and its relationships to other objects in the computer system or network. Such object or node can send and receive messages to and from other objects, respond and react to such messages (e.g. commands) but shall normally be impervious to internal scrutiny. For example, in the above-noted computer data storage system (a kind of computer) each object (system object) may describe or relate to a specific tangible detail in the storage system or in the storage system's processor (e.g., details such as those describing or relating to aspects of operation of the processor's cooling-fan, power switch, cache memory, power supply, disk drive interface, etc.). These tangible objects (nodes) in the storage system can send messages to each other within the storage system and to other objects outside the storage system over the network with which they are operatively coupled. Also, the storage system itself can be an object and interact as a node with other nodes in a network.
The relationship between and amongst these specific objects in the storage system is usually visualized or characterized as a “tree” of objects. In a tree, each such object hangs off a preceding object as if in a parent-child or inheritance relationship, with many children hanging from a parent not being an atypical configuration. In addition to these tangible kinds of objects, logical units (LUNs) are other nodes or objects that can be contained within the tree. For example, a storage system object can have several LUN objects as its children which, in turn, can have various disk objects as their children, etc. These kinds of objects are generically referred to herein as “system objects” since they all relate to a system or to components within a system, whether it is a storage system, computer system, disk drive system, or some other system, and representations of these objects can typically be displayed on a computer terminal's graphical user interface (GUI) in this tree fashion. However, in contrast, other kinds of objects (nodes) can also be formulated which do not relate to a system or its components per se, such as objects relating to user actions and represented on the GUI in other ways. (User actions are any commands or operations initiated by the user, such as, for example, creating a LUN or downloading new software to a disk, etc.). In addition, there can be yet other kinds of network nodes beyond the two types mentioned such as nodes comprising or represented by a communication tree.
Accordingly, it shall be appreciated that there can be a very large number of nodes of various kinds or “personalities” to keep track of and manage within even a relatively small and local computer network. Any one or more of these nodes can fail, for one reason or another, either temporarily or permanently, and certain other nodes can be added or removed by network users under certain conditions, and this complex and dynamic network node scenario must be efficiently and effectively managed if the computer network's intended purpose is to be fulfilled. Moreover, this network node management problem is compounded when the network is large, and possibly worldwide. The subject of these computer network nodes is discussed further in two patent applications filed by the assignee of the present invention: “Plug and Play Interface for User Actions”, Desai et al, U.S. Ser. No. 09/916,102, filed Jul. 26, 2001 and “Scalable Communication Within a Distributed System Using Dynamic Communication Trees”, Bauer et al, U.S. Ser. No. 09/877,862, filed Jun. 8, 2001, both of which are incorporated by reference herein in their respective entireties.
One prior art solution to the problem of managing network nodes is to elect one node to have complete information about all other network nodes in the system. Similarly, a group of nodes can be elected where each elected node in the group is assigned to its own subnet or domain of nodes and has complete information about all other nodes in its own subnet or domain. A subnet or domain is a network unto itself. All other nodes in that network, or in that subnet as the case may be, then seek information necessary to their functioning from their respective one elected node and are thus managed through that one elected node. The inherent weakness in this configuration is that if such one elected node fails, then no other node in its respective network or subnet can function resulting in a failed network or subnet. This is a single point of failure design which is not an optimum design because of at least this problem.
Prior art attempts to make this single point of failure design more reliable included use of backup nodes, to take over functioning of an elected node if and when the elected node failed. The problem with this backup design is that backup nodes can also fail and then the result is the same as before—a failed network or subnet within the network. Furthermore, when dealing with a worldwide network, which is not atypical today, if backup nodes are being relied upon because of potential elected node failures, and if all backup nodes are being maintained in one locale (e.g. United States of America) for convenience, security, or other purposes, then virtually all network node information can be destroyed for that global network if a disaster at the backup locale destroys the backup nodes. This would be a major disruption. All worldwide users (Europeans, Asians, etc.) of that worldwide network would thus lose the service provided by that network.
Effective solutions to these single point of failure problems of the aforementioned prior art are presented in the patent application cross referenced herein, and include use of a master node which can be selected from the ranks of participating nodes in a network domain or configuration. The master node is a privileged node relative to its participating nodes and controls the directory database (DDB) distributed within each of its participating nodes to match its own DDB and thereby be substantially identical from node to node across its domain. The master node is selected or replaced with another node under control of a global administrator, a privileged network user having special password access to enable only him/her to perform these master node changes. It is prudent computer network practice to have only one global administrator per domain, or one person of authority under such global administrator per domain, who selects and changes master nodes.
This master node solution is very satisfactory in any DDB network or domain within that network in which only one master node asserts itself on participating nodes in that network or domain, which may be the usual case, particularly with LANs. However, if another user of the network gains access to the global administrator's password, either with or without proper authorization, and if that other user selects a second master node (a purported or contending master node) in the same domain that had a previously operative master node, then a conflict results. All nodes in the configuration would then be subject to assertions by two different master nodes that each of those master nodes is the true master for each node in that domain. This problem may be compounded further if there are more than only two users having this password access. This problem may be compounded even further if this is a globally-dispersed network where human language barriers can be challenging and result in multiple masters being created by mistake. Further, widely-varying time zone differences (some users are sleeping while others on opposite sides of the globe are working on the global network) might also contribute to multiple masters being created by mistake.
Thus, there is need for a solution to this multiple master conflict problem, and the present invention is an effective solution.
Embodiments of the present invention include apparatus, method, system, and/or computer program product for resolving multiple master node conflict in a computer network having a plurality of nodes only one which should be master node. The conflict is between at least a first purported master node and a second purported master node. One of these nodes is chosen to be master node.
In another aspect of the present invention a computer network has a plurality of nodes each of which has a directory database (DDB) and one of which should be master node. The master node is used to maintain contents of the DDB in each of the nodes consistent throughout the plurality in a manner to avoid a single point of failure. A DDB is a database of directory information, such as a directory of addresses of these nodes. These nodes, such as computer storage systems, interact with computer network information and can receive, store, modify and/or transmit the information. Embodiments of the present invention include apparatus, method, system and/or computer program product to resolve conflict between a first purported or contending master node and a second purported or contending master node within the plurality of nodes, by establishing a standard for comparison between the first purported master node and the second purported master node, by comparing the first purported master node against the second purported master node in accordance with the standard, and by selecting either the first or second purported master node to be the master node based on the comparison. In such apparatus there is a device that establishes the standard, another device that compares the two purported masters, and yet another device that selects one or the other purported master. The purported master node not selected is demoted to non-master status and is returned to the plurality of nodes as a participating node in that plurality.
In yet another aspect of the present invention, the standard is a temporal standard. When the first purported master node was selected master node of the network a first time of selection was obtained and when the second purported master node was selected master node of the network a second time of selection was obtained. The first time and second times of selection are compared to obtain the most recently selected purported master node. The most recently selected purported master node is allowed to be master node and the other purported node is demoted to non-master status as a participating node in the plurality. If both purported master nodes are selected simultaneously, a default condition results whereby the purported master node having the lower IP address becomes the master node. (The higher IP address could have been used instead.)
In a further feature of the present invention, in the first purported master node, the local time of selection of the first purported node as master node is recorded and the duration of the selection is measured to obtain a first selection duration. The duration is communicated to all other nodes in the plurality. Similarly, in the second purported master node, the local time of selection of the second purported node as master node is recorded and the duration of that selection is measured to obtain a second selection duration. The second duration is also communicated to all other nodes in the plurality.
For each one of the nodes in the plurality except for the two purported master nodes (for the slave nodes), the local time of receipt of the first selection duration is noted and the first duration is subtracted from that local time of receipt to obtain first adjusted local time, the local time of receipt of the second selection duration is noted and that second duration is subtracted from that second local time of receipt to obtain second adjusted local time, and the two adjusted local times are compared to obtain the most recent adjusted local time. The node associated with the most recent adjusted local time, either the first purported master node or the second purported master node, is identified in each of the slave nodes.
Concurrently with this activity in the slave nodes, in the first purported master node the local time of arrival of the second selection duration is noted and that duration is subtracted from that local time to obtain a first purported master node adjusted competitive local time. That competitive local time is compared with the first purported master node's local time of selection as master to obtain a first most recent selection time. The node associated with this first most recent selection time, either the first purported master node or the second purported master node, is also identified.
And, concurrently with these activities in the first purported master node and slave nodes, in the second purported master node the local time of arrival of the first selection duration is noted and that duration is subtracted from that local time to obtain a second purported master node adjusted competitive local time. That competitive local time is compared with the second purported master node's local time of selection as master to obtain a second most recent selection time. And the node associated with this second most recent selection time, either the first purported master node or the second purported master node, is also identified.
The number of times that the first purported master node is identified in each of the slave and purported master nodes as the most recently selected master node is tallied to obtain a first total, and the number of times that the second purported master node is identified in each of the slave and purported master nodes as the most recently selected master node is also tallied to obtain a second total. If the first and second totals are equal the tie is broken by selecting the purported master node that has the lower IP address (again, the higher address could have been used as selection criteria). But, if the totals are different, the final master node selection is made by selecting whichever purported master node had the greater total. The purported master node with the lower total is demoted to non-master-node status and returned to the plurality of nodes as a participating node.
It is thus advantageous to employ embodiments of the present invention in any computer network, including client-server networks, whether a LAN or globally dispersed WAN, where management of a distributed directory database is implemented by way of a master-node to slave-node arrangement in which only one master node should exist. For example, it is advantageous to employ embodiments of the present invention in computer networks having a plurality of nodes or storage systems each of which contains a DDB maintained consistent by a master node selected by a global administrator from the plurality, where the maintenance is performed in a manner to avoid a single point of failure and only one master node is permitted for that network at any given time. Embodiments of the present invention ensure that only one master node shall be permitted at any given time.
It is therefore a general object of the present invention to provide an improved computer network.
It is another general object of the present invention to provide improved distributed management software.
It is a further object of the present invention to provide an improved technique for managing a directory database distributed over a plurality of nodes in a network such as a client server network, such technique allowing only one master node at a time in the plurality.
It is an even further object of the present invention to provide an improved computer data storage system.
Other objects and advantages will be understood after referring to the detailed description of the preferred embodiments and to the appended drawings wherein:
Preliminarily, section headings which refer to figures to which they principally relate are used hereinbelow as an organizational aid. However, there may be discussion of subject matter reflected in a particular figure which appears in a section headed by a figure number associated principally with other subject matter. The section headings are thus not intended to be construed in a limiting manner. The terms “master node” and “master” may be used interchangeably herein and have the same meaning. And, it should be kept in mind that any network in which embodiments of the present invention are operative has an intended purpose beyond management of its nodes. For example, in a computer data storage system network, in which embodiments of the present invention are particularly useful, the main purpose of each node or storage system in such network is to enhance the overall network function of storing data, and management of its distributed DDB is only one operation involved in that overall data storage function.
As shown in
These master nodes are selected by a computer network user termed a “global administrator”, a user who is privileged as compared with other users in that computer network. Only the global administrator has proper authorization to select or change master nodes, and make other network changes. Other users have rights limited to merely using the network, as is. The global administrator selects nodes M1, M2, and M3 through a network user interface, e.g. UI 100 of
Assume that network 400 initially was not configured as shown, but that it was merely a boundary which contained nodes 1-8 inclusive in an unconfigured state, such as reflected in
Backtracking momentarily, when the global administrator is adding node 2 to the configuration, such node is necessarily being added through, or by way of, the only node that pre-existed node 2 in the domain 401 configuration, namely node 1. Accordingly, node 1, acting as master node, forwards information to just-arrived node 2 which advises node 2 that node 1 is going to be master for node 2 and that node 2 has been added to the configuration of node 1. Node 2 is thus a “master-added node”. Node 2 then checks the version number of the DDB in information received from the master node. A version number is a unique worldwide number (WWN), a number that if once used is never used again. In this example of a new configuration being created out of a boundary of chaotic or unaligned nodes the version number of the master node's DDB could be zero. If the version number in information sent by the master node is different from the pre-existing version number of the DDB in node 2 (the likely case), then node 2 needs to get a copy of the entire contents of the master's current DDB and to update (replace) the DDB in node 2 to be identical to the DDB in node 1. But, if version numbers of the DDBs in nodes 1 and 2 are the same (not likely), then the foregoing update/replacement is avoided. The same basic procedure is undertaken for node 3 and for any other node to be added to the domain 401 configuration, but there is some significant variation to the procedure as a function of which node in the configuration is selected by the global administrator to be a “portal node”—the node through which a new node is to be added. If the portal node is master node, then the above procedure is used, but if the portal node is a configured node within domain 401 other than master node then a different procedure is used, as follows.
If the global administrator adds node 3 through node 2 thus making node 2 a portal node for new node 3, node 2 stores the IP address of node 3 in a cache memory within node 2 (not within the DDB of node 2), sends a message to master node 1 that node 3 has been or is being added and provides the IP address of node 3 in that message. If master node 1 successfully receives this message from node 2, master node 1 immediately updates itself by adding node 3 IP address information to its DDB and changes its version number. The version number changes from one unique WWN to a different unique WWN. Concurrently, node 1 sends a “success note” back to portal node 2 about which more will be said below. Further, master node 1 sends a message to node 3 that the master for node 3 is node 1 and also provides node 3 with the changed new version number of node 1 as well as a copy of the contents of its entire DDB. Newly arrived node 3 accepts such contents, updates its DDB therewith, applies the changed new version number to such contents, and is thereafter a participating node along with node 2 in a configuration designated as domain 401 in which node 1 is master node.
Although, node 2 has retained the IP address of newly-added node 3 in its cache memory, node 2 has not yet been updated. The earlier-mentioned success note contains at least the master node's prior version number as well as its changed new version number. Node 2 compares the master node's prior version number received in the success note against its own version number of its DDB. On the one hand, if these version numbers match, portal node 2 updates its own DDB with node 3 IP address information which is being stored in the node 2 cache memory, thereby relieving the master node from unnecessarily returning node 3 IP address information back to node 2. Portal node 2 also updates its version number to be the same as the changed new version number of the master node which was received in the success note. But, on the other hand, if the prior version number of master node 1 did not match the version number of portal node 2, then portal node 2 destroys contents in its cache memory about new node 3 and waits for its update message from master node 1. (The update message is a change notification to all participating nodes in the master node's domain and is part of the master node's replication process.) If there were many other operative and participating nodes in domain 401, they would all be subjected to this replication process and portal node 2 would be just one of all such participating nodes being updated with information about newly added node 3 at the time of the update message. A significant aspect of this procedure is that portal node 2 receives minimal communication from master node 1 to achieve update of its DDB as noted above. Accordingly, any master node is worked minimally and network or configuration “housekeeping” traffic is kept minimal in the present invention while maintaining contents of distributed DDBs consistent throughout any given node configuration.
Next, consider node 4, another new node being added to the configuration of domain 401. If the global administrator, by way of the computer network's GUI, selects node 3 as portal node for node 4, and if node 3 cannot properly communicate with master node 1 for one of a number of reasons such as the network link between master and node (not shown in
Domain 402 is shown with nodes 5, 6, 7 and 8. These nodes are shown in a configuration with node 6 as its master node. These four nodes in this configuration have been created in the same manner as those in domain 401, but either by a different global administrator from the one in control of domain 401, or by the same global administrator intending to establish a different and mutually-isolated domain from domain 401. Accordingly, these two domains represent two different configurations which can co-exist in a single computer network without either domain interfering with the other. More than two domains can co-exist in a single network and only two are shown as an example. (Related subjects are further discussed hereinbelow in connection with the section on “Marriage of Two Subnets”). Nodes in this domain are shown to be interconnected by communication lines 403-408 in a particular manner. More is said in the incorporated-by-reference related application about these communication lines and about master node “pinging” and participating node “polling” techniques utilized in certain failure scenarios.
The procedure of adding a new node to a configuration, whether a master or non-master node in the configuration is used as portal node, requires certain handshaking protocols. These protocols shall be discussed in detail hereinbelow in connection with flowcharts depicting their respective algorithms. This discussion of
The algorithmic process then moves to block 503 where the global administrator chooses other nodes through the GUI from either within or outside of the boundary to be added to the configuration. Next, in block 504, the master node sends its IP address to all the other nodes chosen to be within the configuration informing each of them that the master node is their master. (Employing personification, in such “informing” or “announcing” the master is effectively saying: “I am the master for you”.) This informing is the aforementioned pinging and is also used in connection with certain failure scenarios discussed in the incorporated-by-reference application. In block 505 each chosen node enters the master node's IP address in its DDB, acknowledges the master node, and thus aligns itself to the master node. This acknowledgment is vital for the master to receive from each node in its configuration so that the master knows that each such node is properly aligned. Failure to receive an acknowledgment from a particular node alerts the master that such node might have failed and sets up a repetitive pinging activity. The actions in blocks 503, 504, and 505 are accomplished by way of the configure domain dialog box in the GUI. The steps or actions in this Figure are repeated as many times as are necessary to accomplish the total number of domains desired per network. Thus, in
In query block 601, for each node in the configuration, a decision or determination is made regarding whether or not the master node's IP address in the configuration update (for either a node added or node removed update) matches the purported master node's IP address stored in each node's DDB, as verification of the master's identity. Verification is important because participating nodes in a configuration must receive commands and updates from only their one true master node. Without verification a problem could arise. For example, if a master node fails, requiring the global administrator to select a different node as new master node, the new master node sends out notification to all nodes in its configuration that it is their new master node. All participating nodes in the configuration realign themselves to the new master node without problem. But, if the failed master node recovers after selection of the new master node, and if a user or different global administrator erroneously logs-in to that previously failed master, there is a chance that it might be erroneously used as master in the same configuration, which has to be prevented. The verification operation defined by query block 601 prevents this problem from occurring
In block 601, if there is an IP address mismatch whereby the answer is “no” for any particular participating node in the configuration, then in block 605 the change or configuration update is not recognized by that particular node, the event is not logged into the DDB of that particular node, and the handshake is concluded. In effect, this means that the particular node sends a message back to the master node advising it that it is not the particular node's master (in such case the likelihood is that the particular node is participating in a different domain under a different master). Thereupon, the master node removes the IP address of that particular node from the master node's DDB and sends the change to all other nodes in the configuration. However, if the answer is “yes”, the process then moves to query block 602 where version numbers are compared. For each participating node in the configuration, its DDB version number before the configuration update is compared with the master node's DDB version number before the update. This step ensures that each participating node's DDB contents were the same as the master node's DDB contents before the update.
On the one hand, if the version numbers are the same whereby the answer to the query in block 602 is “yes” then in block 604 only the update information (only the IP address of the node being added or the node being removed) is accepted into that participating node's DDB, and the handshake concludes. On the other hand, if the answer is “no” then the version numbers do not match which means that contents of their respective DDBs do not match. There is no readily available technique to determine to what extent any such DDB may be out-of-date. Accordingly, for the situation of a version number mismatch, in block 603 the participating node flushes its contents and accepts the entire contents of the master node's DDB including both updated configuration information about the added or removed node and a changed new master node version number. Thereafter, the handshake concludes.
Referring next to
In block 701, an inquiring node having need to send substantive information to its master gets the name and address of its presumed master stored in its own, local, DDB. In decision block 702 that node inquires of the presumed master if it is that node's master. If the presumed master node responds “yes”, the handshake is concluded. But, if the presumed master node responds “no”, then the algorithmic process moves to another decision block, block 703.
In block 703 the inquiring node inquires if the presumed master knows who is the new master. If the answer is “yes”, where the presumed master knows the identity of the new master, the algorithmic process moves to block 704 wherein the presumed master provides the name and IP address of the new master to the inquiring node after which the handshake is concluded. But, if the answer is “no” the algorithmic process moves to yet another decision block, block 705.
In block 705, which is needed when the presumed master does not know who is the new master, a decision needs to be made about asking the global administrator to configure a new master node by use of the select master dialog box in the GUI (discussed in connection with
With regard to screen layout, in the upper left hand corner of the Figure, a “Select Master” title in the toolbar is shown and in the upper right hand corner in that toolbar an exit button “X” is shown and used to exit the dialog box. Under the heading “Master Node IP Address” there is shown an editable field containing, in this example, a specific IP address: 18.104.22.168. The global administrator can select or change master nodes by editing this field, described below. There is a “Master Candidate Nodes” section in the screenshot. This section displays nodes which are participating, configured nodes in the domain and are thus available as candidates from which one can be selected or elected by the global administrator as master. In this section, three categories of information are provided: “Node Information”, “System Name” and Provider Information”.
With regard to operation, the global administrator can point the cursor and mouse-click on Apply (can hit Apply button) to make a node selected from this group of three nodes appear in the editable field above and become new master for this configuration. In this example, the second candidate in the Master Candidate Nodes section corresponding to Array 2 is selected as master node, as is reflected in the editable field. The OK button performs the same action as the Apply button, but also closes the dialog box. The Cancel button closes the dialog box without saving the global administrator's changes thereby maintaining the status quo. And, the “Help” button launches a help menu for this Select Master dialog box.
However, before any of this operation can be performed, the global administrator first has to bring up the dialog box. And, before the dialog box can be brought up, the global administrator first has to bring up a web browser on his/her terminal screen and type into the Uniform Resource Locator (URL) slot the IP address of a storage system chosen by the global administrator (in this example, one of the IP addresses for Array 1, 2, or 3). Such storage system may be part of a small or large domain or network of storage systems and possibly other nodes. In this example the global administrator could have typed-in the IP address of any of the three arrays shown and the same substantive information would have been displayed because the DDB in each of the three storage systems contains the same directory data. However, the information may be displayed in a different order as a function of which IP address was typed in. By accessing a system or node in this manner, a framework software is brought into operation, which provides the framework or foundation of the terminal screen display. With reference to one of the incorporated-by-reference patent applications, “Plug and Play Interface for User Actions”, Desai et al, U.S. Ser. No. 09/916,102, filed Jul. 26, 2001, a dialog box of this kind can be created by a software module that plugs into such framework software as described in that patent application. However, after framework software comes up, a security operation involving password access is next undertaken to ensure that only a user with proper password (presumably global administrator) has access to the Select Master dialog box. If the proper password is supplied, the framework software thereafter offers a menu item “Select Master”. And, if that menu item is pointed-to and clicked-on by the global administrator, or other authorized user, then, this dialog box is finally launched.
The act of launching this dialog box causes reading of directory information from the DDB in the storage system node chosen by the global administrator. In this example, that DDB contains node, system name, and provider information of Array 3, Array 2 and Array 1 respectively, and displays it as shown. This information shows that this DDB has three storage arrays—three IP addresses in this case—and that the chosen storage system is thus participating in a domain or configuration that has three nodes. In the case where a storage array has two IP addresses associated with it, the same system name will appear on the screen twice, reflecting both IP addresses on two separate rows.
Any one of these three nodes (if all are operative) can be selected by the global administrator to be master node by merely highlighting that selection in the Master Node Candidates section and hitting the OK or Apply button. This action will move the IP address corresponding to that selection to the Master Node IP Address editable field and replace the IP address of any node that might have been in that field. That replaced master node is then demoted to being a participating node of the configuration. (Referring back to
With regard to screen layout, there are three sections displayed: “Domain Name” near the top, “Scan Subnets” below that, and “Select Systems” at the bottom half of the screenshot.
With regard to operation, this dialog is brought up in the same manner as described earlier in connection with the dialog of
In the Subnets to Add edit control the global administrator types in the address of a subnet which he/she wishes to have scanned to discover any available systems (systems unaligned to any master node) that are also compatible with other nodes in the domain which the global administrator is in the process of configuring or re-configuring. For example, nodes or storage systems or servers which fall into the category of Common Information Model Object Manager (CIMOM) systems or servers are mutually compatible. Note that this address is not a full four-segment IP address, but is a subnet address or identifier (ID) having only three segments. The “A” button located to the right of the Subnets To Add edit control is used to for moving contents of that edit control to the Subnets to Scan list control located to the right of the button, thereby populating that list control field which is initially empty when the dialog is brought up. A populated list control field thus reflects certain subnets which the global administrator wishes to scan. As can be seen, in this example subnet address “10.14.12” is the last subnet that was added to the Subnets to Scan field, as it is located at bottom of three subnet addresses shown in that field. The global administrator uses the screen cursor to highlight one or more subnets listed and hits the Scan button to the right of the list control to start a discovery operation on those highlighted subnets. The discovery operation will cause a display in the progress bar to show percentage of completion of scan operation. The Scan button is disabled from time that scanning is started until it is finished. The Clear button is used for clearing contents of list control, this button being enabled only if the global administrator selects (highlights) a subnet listed in the list control. The Clear button is disabled and a subnet cannot be highlighted in the list control field when scan operation is in progress. The Stop Scan button is used to stop scan operation which was started by hitting the Scan button, and will be enabled only if there is a scan in progress.
After a scanning operation has completed, where progress bar shows 100%, IP addresses of any discovered nodes that are unaligned to second purported master node and which are compatible with nodes in the domain under configuration are automatically sent to the Available Systems field in the Select Systems section of the dialog. In other words, these particular discovered nodes populate the Available Systems field which has sub headings of “System” (storage system name) and “Node Info” (storage system IP address). In the example shown, in the Subnets To Scan field, the subnet having address 128.221.34 contained an unaligned and compatible node with IP address 22.214.171.124 with system name Cadsys, and this is the first entry in the Available Systems window. Likewise, the subnet having address 128.221.42 contained three unaligned and compatible nodes: 126.96.36.199; 188.8.131.52; and, 184.108.40.206 which are three entries under Node Info in the Available Systems field. Finally, the last subnet scanned having address 10.14.12 contained an unaligned and compatible node with IP address of 10.14.12.105 which is the last entry shown in the Available Systems field.
The global administrator now has a choice of selecting any or all of these available systems for addition to the domain being configured. For example, to add the node named Cadsys the global administrator would highlight this entry in the Available Systems field and hit the C button. This action would move that Cadsys entry from the Available Systems field which would thus be decreased by one entry, to the Selected Systems field to the right of the button which would thus be increased by one entry. Then the global administrator hits the Apply or OK button to cause the master node to replicate the node added change in the DDBs of all nodes in the domain and thereby keep the DDBs consistent throughout the domain.
In the Selected Systems field there are shown three nodes having the same IP addresses that were shown in
In the reverse operation, the population of the Selected Systems field can be reduced by the global administrator highlighting an entry in the Selected Systems field and hitting button “D” which moves that highlighted entry into the Available Systems field. Then the global administrator hits the Apply button which causes the DDB in the removed node to destruct (to discard its contents), thereby unaligning its DDB from alignment to master node cpc34187. This unalignment makes the removed node available for use by other global administrators. Hitting the Apply button also causes the master node to replicate this node removal change in all the remaining DDBs in the other nodes of this domain and thereby keep the DDBs consistent throughout the domain. The “Clear” button is used to clear or empty contents of the Available Systems field to make it open for repopulation by newly-discovered available systems derived from the scanning operation in the Scan Subnets section of the dialog. The cleared systems which are thus removed from view from the Available Systems field are otherwise unaffected and their DDBs remain intact. Scroll bars at bottom of both the Available Systems field and the Selected Systems field, also operable through left and right pointing arrow buttons at lower left and right corners of the fields, permit vertical scrolling of entries in both fields.
In the editable “IP Address of System” field located above the Available Systems field the IP address 220.127.116.11 is shown. This is an IP address of a node about which the global administrator has prior knowledge, and knows is both unaligned to any other master and compatible with nodes of this domain under configuration. The global administrator has thus typed-in this IP address into this editable field. By clicking on the “B” button, this IP address is moved into the selected Systems field and the OK or Apply button is then hit to add the node with this IP address to the configuration as described above. If the global administrator later wants to remove it, highlighting it and hitting the D button moves it over to the Available Systems field, in accordance with operation discussed above.
It should be understood that embodiments of the present invention relating to the GUI discussion of
With the GUI operational description of
The result of this instant scan will cause node 2 to initiate an update to its master node 1, advising it of new nodes in the domain. Master node 1 will update its DDB with all node information from all nodes in both subnets. But, when master node 1 propagates its update to all nodes from both subnets, those nodes which comprise domain 402 reject the update. Those nodes have their own master node, master node 6. Because of the master to node handshake, nodes 5, 6, 7, and 8 will reject this update request since the IP address of master node 6 differs from the IP address of master node 1 (see
This network configuration has practical utility when a link between domains is useful or economically feasible, but not on a continuous basis. For one example, consider the case where a startup company on a tight budget has two offices located on opposite coasts of the United States, and where they have need to access each others files. This business arrangement might arise when a West coast project is started at a different time from the East coast project and where it is useful to keep the two projects segregated except for intra-company access to each other's files. This company may lease network lines that are charged at an hourly rate to connect these two subnets cross country. And, with this marriage scheme, they can be disconnected each evening and reconnected each morning to effect a substantial savings in leased line costs. While these subnets are connected, co-workers at opposite ends of the country can read each others files at great convenience to the company, but they cannot otherwise impact each other's files, keeping the two projects segregated. The DDBs in the East coast's nodes are isolated from the West coast's master node and vice versa.
In the example shown, a global administrator located in the USA properly selects a node located in Brazil to be master node of this global network. This selection is made via communication link 1005 by way of the dialog of
The global administrator, as noted above, is a network user with special privileges. Only this person, or someone under his/her authority, has appropriate password access to the dialog of
When master node M-2 had been selected, in addition to its announcement as master, it had recorded its local Brazilian time of selection (LTS) as master and thereupon starts recording its time duration as master node. Analogously and figuratively speaking, master node M-2 starts its “stop watch” when it is selected as this network's master. Likewise, when unauthorized master node M-1 is selected, in addition to its announcement as master, it records its local Japanese time of selection as master and thereupon starts recording its time duration as master node. In addition to the announcements made to the other nodes in the network by these masters at the times of their respective selections (appointments), each also transmits its respective time duration as master node to all other nodes in the network. These transmissions of time duration are not necessarily coincident with the announcements, and any time delay from time of selection until transmission of the time duration is accounted for in the transmitted time duration value itself.
For example, consider node 4 which is shown with links to both master nodes. Brazilian Master node M-2 sends its time duration value to node 4 in the USA on transmission link 1002 and node 4 subtracts that value from its own local time to achieve an adjusted local time of selection of M-2 as master node (shown as NODE 4's ADJ. LTS-2). Node 4 records this LTS-2 time. Similarly, Japanese master node M-1 sends its time duration value to node 4 in the USA on transmission link 1001 and node 4 also subtracts that value from its own local time to achieve an adjusted local time of selection of M-1 as master node (shown as NODE 4's ADJ. LTS-1). Again, node 4 records this LTS-1 time. As noted, this process (this time duration transmission and subtraction from local time process) involves all nodes in the network. Accordingly, each master node transmits to the other master node over transmission link 1003 its respective time duration. At node M-1, it receives M-2's time duration which is subtracted from local Japanese time to become Node M-1's adjusted LTS-2 which is stored in M-1. Likewise, at node M-2, it receives M-1's time duration which is subtracted from local Brazilian time to become node M-2's adjusted LTS-1 which is stored in M-2. Each node in the global network is thus supplied with temporal information with which it can make a meaningful comparison in its respective local time zone between the absolute time of selection of M-1 and the absolute time of selection of M-2. The detail with respect to this comparison is provided in connection with discussion of flowcharts in
The present embodiments are to be considered in all respects as illustrative and not restrictive. The flowcharts used herein to demonstrate various aspects of the invention should not be construed to limit the present invention to any particular logic flow or logic implementation. For example, the alternative embodiment of the master to node handshake as depicted in
The present invention may be used in any network environment where a distributed directory database is needed or utilized, such network being used for any purpose including, but not limited to, computer data storage. Furthermore, although embodiments of the present invention include C++ and JAVA object oriented software, other software could be utilized. Therefore, the scope of the invention is indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are intended to be embraced therein.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US5289176 *||Mar 19, 1992||Feb 22, 1994||Aeg Transportation Systems, Inc.||Multi-master resolution of a serial bus|
|US5313465 *||May 13, 1992||May 17, 1994||Digital Equipment Corporation||Method of merging networks across a common backbone network|
|US5511188 *||Dec 30, 1993||Apr 23, 1996||Johnson Service Company||Networked facilities management system with time stamp comparison for data base updates|
|US5608903 *||Dec 15, 1994||Mar 4, 1997||Novell, Inc.||Method and apparatus for moving subtrees in a distributed network directory|
|US5708812 *||Jan 18, 1996||Jan 13, 1998||Microsoft Corporation||Method and apparatus for Migrating from a source domain network controller to a target domain network controller|
|US5768519 *||Jan 18, 1996||Jun 16, 1998||Microsoft Corporation||Method and apparatus for merging user accounts from a source security domain into a target security domain|
|US5805072 *||Dec 1, 1995||Sep 8, 1998||Ultra-High Speed Network||VC connection method|
|US5805785 *||Feb 27, 1996||Sep 8, 1998||International Business Machines Corporation||Method for monitoring and recovery of subsystems in a distributed/clustered system|
|US5805786 *||Jul 23, 1996||Sep 8, 1998||International Business Machines Corporation||Recovery of a name server managing membership of a domain of processors in a distributed computing environment|
|US5844900 *||Sep 23, 1996||Dec 1, 1998||Proxim, Inc.||Method and apparatus for optimizing a medium access control protocol|
|US5968121 *||Aug 13, 1997||Oct 19, 1999||Microsoft Corporation||Method and apparatus for representing and applying network topological data|
|US5974429 *||Sep 3, 1997||Oct 26, 1999||Northern Telecom Limited||Method and apparatus for updating distributed databases in a telecommunications network|
|US5987011 *||Aug 30, 1996||Nov 16, 1999||Chai-Keong Toh||Routing method for Ad-Hoc mobile networks|
|US6061600 *||May 9, 1997||May 9, 2000||I/O Control Corporation||Backup control mechanism in a distributed control network|
|US6081898 *||Aug 5, 1997||Jun 27, 2000||Microsoft Corporation||Unification of directory service with file system service|
|US6092214 *||Nov 6, 1997||Jul 18, 2000||Cisco Technology, Inc.||Redundant network management system for a stackable fast ethernet repeater|
|US6115713 *||Aug 30, 1996||Sep 5, 2000||Johnson Controls Technology Company||Networked facilities management system|
|US6181689 *||Oct 23, 1996||Jan 30, 2001||Ncr Corporation||Bi-directional information exchange mechanism for collaborative network navigation among a group of user terminals|
|US6199169 *||Dec 15, 1998||Mar 6, 2001||Compaq Computer Corporation||System and method for synchronizing time across a computer cluster|
|US6243826 *||Jun 22, 2000||Jun 5, 2001||Cisco Technology, Inc.||Redundant network management system for a stakable fast ethernet repeater|
|US6292905 *||Oct 2, 1997||Sep 18, 2001||Micron Technology, Inc.||Method for providing a fault tolerant network using distributed server processes to remap clustered network resources to other servers during server failure|
|US6295541 *||Aug 18, 1998||Sep 25, 2001||Starfish Software, Inc.||System and methods for synchronizing two or more datasets|
|US6324571 *||Sep 21, 1998||Nov 27, 2001||Microsoft Corporation||Floating single master operation|
|US6363416 *||Aug 28, 1998||Mar 26, 2002||3Com Corporation||System and method for automatic election of a representative node within a communications network with built-in redundancy|
|US6381627 *||Sep 21, 1998||Apr 30, 2002||Microsoft Corporation||Method and computer readable medium for discovering master DNS server computers for a given domain name in multiple master and multiple namespace configurations|
|US6450298 *||Aug 16, 2001||Sep 17, 2002||Mitsubishi Denki Kabushiki Kaisha||Communications control system for elevators|
|US6466608 *||Apr 6, 1995||Oct 15, 2002||Proxim, Inc.||Frequency hopping medium access control protocol for a communication system having distributed synchronization|
|US6611860 *||Nov 17, 1999||Aug 26, 2003||I/O Controls Corporation||Control network with matrix architecture|
|US6636982 *||Mar 3, 2000||Oct 21, 2003||International Business Machines Corporation||Apparatus and method for detecting the reset of a node in a cluster computer system|
|US6665730 *||Dec 16, 1999||Dec 16, 2003||At&T Corp.||Method and apparatus for transaction routing in a connection-oriented packet network using a non-fault-tolerant directory server|
|US6680903 *||Jul 7, 1999||Jan 20, 2004||Matsushita Electric Industrial Co., Ltd.||Network system, network terminal, and method for specifying location of failure in network system|
|US6691173 *||Jul 6, 1999||Feb 10, 2004||Widcomm, Inc.||Distributed management of an extended network containing short-range wireless links|
|US6732202 *||Nov 17, 1999||May 4, 2004||I/O Controls Corporation||Network node with plug-in identification module|
|US6751634 *||Aug 26, 1999||Jun 15, 2004||Microsoft Corporation||Method and system for detecting object inconsistency in a loosely consistent replicated directory service|
|US6751719 *||Oct 26, 2000||Jun 15, 2004||International Business Machines Corporation||Method and an apparatus to dynamically order features and to resolve conflicts in a multiple-layer logical volume management environment|
|US6823429 *||Sep 10, 1999||Nov 23, 2004||International Business Machines Corporation||Memory controller for controlling memory accesses across networks in distributed shared memory processing systems|
|US6829487 *||Mar 18, 2002||Dec 7, 2004||Nokia Corporation||Method and apparatus for establishing a communication group|
|US6877043 *||Apr 4, 2001||Apr 5, 2005||Broadcom Corporation||Method for distributing sets of collision resolution parameters in a frame-based communications network|
|US6915309 *||Jun 20, 2001||Jul 5, 2005||Cisco Technology, Inc.||Automatically generating replication topology information for use by a directory service|
|US6941350 *||Oct 19, 2000||Sep 6, 2005||International Business Machines Corporation||Method and apparatus for reliably choosing a master network manager during initialization of a network computing system|
|US7054910 *||Dec 20, 2001||May 30, 2006||Emc Corporation||Data replication facility for distributed computing environments|
|US7120690 *||Sep 27, 2001||Oct 10, 2006||Emc Corporation||Managing a distributed directory database|
|US20020044549 *||Jun 8, 2001||Apr 18, 2002||Per Johansson||Efficient scatternet forming|
|US20020080807 *||Dec 22, 2000||Jun 27, 2002||Lind Carina Maria||Systems and methods for queue-responsible node designation and queue-handling in an IP network|
|US20030149735 *||Jun 22, 2001||Aug 7, 2003||Sun Microsystems, Inc.||Network and method for coordinating high availability system services|
|US20050055418 *||Oct 29, 2001||Mar 10, 2005||Sun Microsystems Inc||Method to manage high availability equipments|
|US20050063313 *||Jul 3, 2001||Mar 24, 2005||International Business Machines Corporation||Optimal clustering of master-slave ad-hoc wireless network|
|US20050102704 *||Nov 7, 2003||May 12, 2005||Rudy Prokupets||Multiregional security system integrated with digital video recording and archiving|
|US20050114401 *||Nov 17, 2003||May 26, 2005||Conkel Dale W.||Enterprise directory service domain controller replication alert and repair|
|US20050163133 *||Jan 21, 2005||Jul 28, 2005||Hopkins Samuel P.||Method for optimally utilizing a peer to peer network|
|US20050182942 *||Feb 17, 2004||Aug 18, 2005||Doru Calin||Methods and devices for obtaining domain access rights|
|US20070081535 *||Oct 4, 2006||Apr 12, 2007||Huawei Technologies Co., Ltd.||Method and system for implementing virtual router redundacy protocol on a resilient packet ring|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US7515546 *||Dec 19, 2001||Apr 7, 2009||Alcatel-Lucent Canada Inc.||Method and apparatus for automatic discovery of network devices with data forwarding capabilities|
|US7624126 *||Jun 25, 2003||Nov 24, 2009||Microsoft Corporation||Registering for and retrieving database table change information that can be used to invalidate cache entries|
|US7720008 *||Aug 12, 2005||May 18, 2010||Siemens Aktiengesellschaft||Method for determining a leading subscriber in a network|
|US7730180 *||Jul 20, 2007||Jun 1, 2010||Emc Corporation||Resolving multiple master node conflict in a DDB|
|US7738504||Dec 22, 2008||Jun 15, 2010||The United States Of America As Represented By The Director National Security Agency||Method of establishing and updating master node in computer network|
|US7969919 *||Aug 8, 2005||Jun 28, 2011||Rockwell Collins, Inc.||System and method for thermal load sharing between nodes in a communications network|
|US7975016 *||Oct 29, 2001||Jul 5, 2011||Oracle America, Inc.||Method to manage high availability equipments|
|US7996093 *||Jan 24, 2008||Aug 9, 2011||Rockwell Automation Technologies, Inc.||Automatic controller relationship resolution|
|US8290919||Aug 27, 2010||Oct 16, 2012||Disney Enterprises, Inc.||System and method for distributing and accessing files in a distributed storage system|
|US8392368 *||Aug 27, 2010||Mar 5, 2013||Disney Enterprises, Inc.||System and method for distributing and accessing files in a distributed storage system|
|US8504684 *||Jun 23, 2010||Aug 6, 2013||Emc Corporation||Control of data storage system management application activation|
|US8543546 *||Sep 10, 2012||Sep 24, 2013||Infoblox Inc.||Semantic replication|
|US8671151 *||Jan 24, 2007||Mar 11, 2014||Oracle International Corporation||Maintaining item-to-node mapping information in a distributed system|
|US8767690 *||Feb 22, 2010||Jul 1, 2014||Sony Corporation||Wireless communication device, wireless communication system, wireless communication method, and program|
|US8768981||Aug 27, 2010||Jul 1, 2014||Disney Enterprises, Inc.||System and method for distributing and accessing files in a distributed storage system|
|US8788639 *||May 13, 2005||Jul 22, 2014||Hewlett-Packard Development Company, L.P.||Method and apparatus for centrally configuring network devices|
|US8875039 *||Nov 18, 2003||Oct 28, 2014||The Mathworks, Inc.||Propagation of characteristics in a graphical model environment|
|US8892516||Jun 27, 2013||Nov 18, 2014||Infoblox Inc.||Provisional authority in a distributed database|
|US8903764 *||Apr 25, 2012||Dec 2, 2014||International Business Machines Corporation||Enhanced reliability in deduplication technology over storage clouds|
|US9011247||Jul 30, 2010||Apr 21, 2015||Wms Gaming, Inc.||Controlling casino lighting content and audio content|
|US9026493 *||Feb 28, 2011||May 5, 2015||Google Inc.||Multi-master RDBMS improvements for distributed computing environment|
|US9203899 *||Sep 29, 2011||Dec 1, 2015||Infoblox Inc.||Event management|
|US9229819 *||Oct 27, 2014||Jan 5, 2016||International Business Machines Corporation||Enhanced reliability in deduplication technology over storage clouds|
|US9317545||Jul 31, 2013||Apr 19, 2016||Infoblox Inc.||Transactional replication|
|US9367987||Apr 26, 2011||Jun 14, 2016||Bally Gaming, Inc.||Selecting color in wagering game systems|
|US9378292 *||Jun 5, 2009||Jun 28, 2016||International Business Machines Corporation||Support for multiple component editions|
|US9520018||Feb 4, 2015||Dec 13, 2016||Bally Gaming, Inc.||Controlling priority of wagering game lighting content|
|US9537761 *||Jul 3, 2013||Jan 3, 2017||Telefonaktiebolaget L M Ericsson (Publ)||IP address allocation in split brain ICR scenario|
|US9547952||Jul 29, 2014||Jan 17, 2017||Bally Gaming, Inc.||Presenting lighting content in wagering game systems|
|US20030112765 *||Dec 19, 2001||Jun 19, 2003||Alcatel Canada Inc.||Method and apparatus for automatic discovery of network devices with data forwarding capabilities|
|US20030163780 *||Jan 29, 2003||Aug 28, 2003||Marc Kossa||Enhancing management of a distributed computer system|
|US20040267824 *||Jun 25, 2003||Dec 30, 2004||Microsoft Corporation||Registering for and retrieving database table change information that can be used to invalidate cache entries|
|US20050055418 *||Oct 29, 2001||Mar 10, 2005||Sun Microsystems Inc||Method to manage high availability equipments|
|US20050107998 *||Nov 18, 2003||May 19, 2005||The Mathworks, Inc.||Propagation of characteristics in a graphical model environment|
|US20050144320 *||Dec 16, 2003||Jun 30, 2005||Jagana Venkata R.||Avoiding conflict of a mobile node IP address|
|US20060256735 *||May 13, 2005||Nov 16, 2006||Hewlett Packard Company Intellectual Property Administration||Method and apparatus for centrally configuring network devices|
|US20080069001 *||Aug 12, 2005||Mar 20, 2008||Rudolf Aschauer||Method For Determining A Leading Subscriber In A Network|
|US20080091802 *||Oct 13, 2006||Apr 17, 2008||Wen-Jea Whan||Plug and play scheme for IP based devices and their failover scheme for quality of services|
|US20080177741 *||Jan 24, 2007||Jul 24, 2008||Oracle International Corporation||Maintaining item-to-node mapping information in a distributed system|
|US20090192645 *||Jan 24, 2008||Jul 30, 2009||Rockwell Automation Technologies, Inc.||Automatic controller relationship resolution|
|US20090307709 *||Jun 5, 2009||Dec 10, 2009||Glenn Rasmussen||Support For Multiple Component Editions|
|US20100272083 *||Feb 22, 2010||Oct 28, 2010||Sony Corporation||Wireless communication device, wireless communication system, wireless communication method, and program|
|US20120023195 *||Sep 29, 2011||Jan 26, 2012||Infoblox Inc.||Event management|
|US20130060734 *||Sep 10, 2012||Mar 7, 2013||Infoblox Inc.||Semantic replication|
|US20130290274 *||Apr 25, 2012||Oct 31, 2013||International Business Machines Corporation||Enhanced reliability in deduplication technology over storage clouds|
|US20140025634 *||Jul 11, 2013||Jan 23, 2014||International Business Machines Corporation||Generating database sequences in a replicated database environment|
|US20140313881 *||Jul 3, 2013||Oct 23, 2014||Telefonaktiebolaget L M Ericsson (Publ)||Ip address allocation in split brain icr scenario|
|US20150046410 *||Oct 27, 2014||Feb 12, 2015||International Business Machines Corporation||Enhanced reliability in deduplication technology over storage clouds|
|US20150180714 *||Dec 24, 2013||Jun 25, 2015||International Business Machines Corporation||Configuration updates across peer storage systems|
|EP2193635A1 *||Sep 26, 2008||Jun 9, 2010||Allied Telesis Holdings K.K.||Method and apparatus for preventing network conflict|
|EP2193635A4 *||Sep 26, 2008||Aug 31, 2011||Allied Telesis Holdings Kk||Method and apparatus for preventing network conflict|
|U.S. Classification||709/224, 709/223, 707/E17.032|
|Sep 27, 2001||AS||Assignment|
Owner name: EMC CORPORATION, MASSACHUSETTS
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:KRISHNAN, SRIRAM;LAZAR, GREGORY W.;REEL/FRAME:012215/0011
Effective date: 20010925
|Mar 11, 2011||FPAY||Fee payment|
Year of fee payment: 4
|Mar 11, 2015||FPAY||Fee payment|
Year of fee payment: 8
|Sep 21, 2016||AS||Assignment|
Owner name: CREDIT SUISSE AG, CAYMAN ISLANDS BRANCH, AS COLLAT
Free format text: SECURITY AGREEMENT;ASSIGNORS:ASAP SOFTWARE EXPRESS, INC.;AVENTAIL LLC;CREDANT TECHNOLOGIES, INC.;AND OTHERS;REEL/FRAME:040134/0001
Effective date: 20160907
Owner name: THE BANK OF NEW YORK MELLON TRUST COMPANY, N.A., A
Free format text: SECURITY AGREEMENT;ASSIGNORS:ASAP SOFTWARE EXPRESS, INC.;AVENTAIL LLC;CREDANT TECHNOLOGIES, INC.;AND OTHERS;REEL/FRAME:040136/0001
Effective date: 20160907
|Sep 29, 2016||AS||Assignment|
Owner name: EMC IP HOLDING COMPANY LLC, MASSACHUSETTS
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:EMC CORPORATION;REEL/FRAME:040203/0001
Effective date: 20160906